You are here

class SnowballStemmer in Snowball Stemmer 2.x

Same name and namespace in other branches
  1. 8 src/Plugin/search_api/processor/SnowballStemmer.php \Drupal\snowball_stemmer\Plugin\search_api\processor\SnowballStemmer

Stems search terms.

Plugin annotation


@SearchApiProcessor(
  id = "snowball_stemmer",
  label = @Translation("Snowball stemmer"),
  description = @Translation("Stems search terms, using <a href=""https://github.com/wamania/php-stemmer"">PHP snowball stemmer</a>, which supports multiple languages. For best results, use after tokenizing."),
  stages = {
    "pre_index_save" = 0,
    "preprocess_index" = 0,
    "preprocess_query" = 0,
  }
)

Hierarchy

  • class \Drupal\snowball_stemmer\Plugin\search_api\processor\SnowballStemmer extends \Drupal\search_api\Plugin\search_api\processor\Stemmer

Expanded class hierarchy of SnowballStemmer

1 file declares its use of SnowballStemmer
StemmerTest.php in tests/src/Unit/Plugin/Processor/StemmerTest.php

File

src/Plugin/search_api/processor/SnowballStemmer.php, line 25

Namespace

Drupal\snowball_stemmer\Plugin\search_api\processor
View source
class SnowballStemmer extends StemmerProcessor {

  /**
   * The stemmer service.
   *
   * @var \Drupal\snowball_stemmer\Stemmer|null
   */
  protected $stemmer;

  /**
   * The language manager.
   *
   * @var \Drupal\Core\Language\LanguageManagerInterface
   */
  protected $languageManager;

  /**
   * {@inheritdoc}
   */
  public static function create(ContainerInterface $container, array $configuration, $plugin_id, $plugin_definition) {

    /** @var static $processor */
    $processor = parent::create($container, $configuration, $plugin_id, $plugin_definition);
    $processor
      ->setStemmer($container
      ->get('snowball_stemmer.stemmer'));
    $processor
      ->setLanguageManager($container
      ->get('language_manager'));
    return $processor;
  }

  /**
   * Retrieves the stemmer service.
   *
   * @return \Drupal\snowball_stemmer\Stemmer
   *   The stemmer service.
   */
  public function getStemmer() {
    $stemmer = $this->stemmer ?: \Drupal::service('snowball_stemmer.stemmer');

    // @todo allow multilingual overrides.
    $stemmer
      ->setOverrides($this->configuration['exceptions']);
    return $stemmer;
  }

  /**
   * Sets the stemmer service.
   *
   * @param \Drupal\snowball_stemmer\Stemmer $stemmer
   *   The stemmer service.
   *
   * @return $this
   */
  public function setStemmer(Stemmer $stemmer) {
    $this->stemmer = $stemmer;
    return $this;
  }

  /**
   * Retrieves the language manager.
   *
   * @return \Drupal\Core\Language\LanguageInterface
   *   The language manager.
   */
  public function getLanguageManager() {
    return $this->languageManager ?: \Drupal::service('language_manager');
  }

  /**
   * Sets the language manager.
   *
   * @param \Drupal\Core\Language\LanguageManagerInterface $language_manager
   *   The language manager.
   *
   * @return $this
   */
  public function setLanguageManager(LanguageManagerInterface $language_manager) {
    $this->languageManager = $language_manager;
    return $this;
  }

  /**
   * {@inheritdoc}
   */
  public function preprocessIndexItems(array $items) {
    foreach ($items as $item) {
      if ($this
        ->getStemmer()
        ->setLanguage($item
        ->getLanguage())) {
        foreach ($item
          ->getFields() as $name => $field) {
          if ($this
            ->testField($name, $field)) {
            $this
              ->processField($field);
          }
        }
      }
    }
  }

  /**
   * {@inheritdoc}
   */
  public function preprocessSearchQuery(QueryInterface $query) {

    // Try the queries language, and if not fallback to the current site
    // language.
    if ($languages = $query
      ->getLanguages()) {

      // Todo: What if there is more than one language?
      $language = reset($languages);
    }
    else {
      $language = $this
        ->getLanguageManager()
        ->getCurrentLanguage()
        ->getId();
    }
    $keys =& $query
      ->getKeys();
    if (isset($keys) && $this
      ->getStemmer()
      ->setLanguage($language)) {
      $this
        ->processKeys($keys);
    }
    $conditions = $query
      ->getConditionGroup();
    $this
      ->processConditions($conditions
      ->getConditions());
  }

  /**
   * {@inheritdoc}
   */
  protected function process(&$value) {

    // In the absence of the tokenizer, and/or HTML processor, this ensures
    // split words for stemming. Leaves strings in much the same state as
    // search api will for storage.
    $words = preg_split('/[^\\p{L}\\p{N}]+/u', strip_tags($value), -1, PREG_SPLIT_NO_EMPTY);
    $stemmed = [];
    foreach ($words as $word) {
      $stemmed[] = $this
        ->stem($word);
    }
    $value = implode(' ', $stemmed);
  }

  /**
   * Actually stem word, if required.
   */
  protected function stem($word) {
    if (strlen($word)) {
      $word = $this
        ->getStemmer()
        ->stem($word);
    }
    return $word;
  }

}

Members

Namesort descending Modifiers Type Description Overrides
SnowballStemmer::$languageManager protected property The language manager.
SnowballStemmer::$stemmer protected property The stemmer service.
SnowballStemmer::create public static function
SnowballStemmer::getLanguageManager public function Retrieves the language manager.
SnowballStemmer::getStemmer public function Retrieves the stemmer service.
SnowballStemmer::preprocessIndexItems public function
SnowballStemmer::preprocessSearchQuery public function
SnowballStemmer::process protected function
SnowballStemmer::setLanguageManager public function Sets the language manager.
SnowballStemmer::setStemmer public function Sets the stemmer service.
SnowballStemmer::stem protected function Actually stem word, if required.